Deriving Culturally Competent HIV Prevention Models Mexican American Injection Drug Users

Antonio L. Estrada, Ph.D., M.S.P.H.


Mexican American Studies and Research Center, University of Arizona, Tucson, Arizona

I. Introduction

As the HIV epidemic enters its second decade in the United States, Hispanics continue to show increasing rates of HIV infection than non-Hispanic Whites (Bakeman, et al., 1987; Peterson and Bakeman, 1989; Curran, et al., 1988; Selik et al., 1989; CDC, 1992). However, given the heterogeneity within the Hispanic population in terms of education, income and ethnic subgroup, it is very important to differentiate levels of risk for acquiring HIV among the various Hispanic subgroups (e.g. Mexican American, Puerto Rican, Cuban, Central/South American) in order to specifically target these groups along a continuum of prevention including primary, secondary, and tertiary prevention components.

The present paper will focus on Mexican-origin IDUs in order to explicate the above variations, describe prevention models which appear applicable, and highlight cultural factors which should be considered in developing culturally competent prevention models for this specific Hispanic subgroup. The focus of the paper is on male injection drug users. Although some cultural concepts can be applied to female IDUs, there are specific cultural nuances of female IDUs that are not described here.

A. Epidemiology of AIDS Among Mexican-origin IDUs in the Southwest

The epidemiology of AIDS among Hispanics (who are primarily Mexican-origin) residing in the five southwestern states of Arizona, California, Colorado, New Mexico, and Texas shows some variation in reference to the transmission categories of IV drug use (COSSMHO, 1991). In Arizona, Hispanics represented 9.9% of AIDS cases. Among these cases, 13.3% were linked to intravenous drug use and 12.5% were linked to male homosexual/bisexual contact and intravenous drug use. In California, Hispanics represented 13.2% of AIDS cases. Of these cases, 6.9% were linked to IV drug use and 7.2% were linked to male homosexual/bisexual contact and IV drug use. In Colorado, Hispanics represented 9.6% of AIDS cases. Of these cases, 10.5% were linked to IV drug use and 13.1% were linked to male homosexual/bisexual contact and IV drug use. In New Mexico, Hispanics represented 28.4% of AIDS cases. Of these cases, 9.1% were linked to IV drug use, and 10.1% were linked to both male homosexual/bisexual contact and intravenous drug use. In Texas, Hispanics represent 13.5% of AIDS cases. Of these cases, 7.6% were linked to IV drug use, and 7.7% of cases were linked to male homosexual/bisexual contact and IV drug use.

When the transmission categories of IV drug use and IV drug use/ homosexual/bisexual contact are combined, the proportion of Hispanics infected through these modes of transmission ranges from a low of 14.1% in California to a high of 25.8% in Arizona.

In order to effectively target a specific Hispanic subgroup for HIV risk reduction, prevention specialists must have a clear idea of the regional variations in the epidemiology of drugs used as well as an epidemiological profile of behaviors which place each subgroup at risk. Unfortunately, few studies document these variations.

B. Patterns of Drug Use Among Hispanic IDUs

An important consideration in the prevention of HIV transmission is an assessment of the types of drugs injected since this has implications on the frequency of injection given each drug's pharmacological effects. Estrada (1991) compared Mexican-American IDUs to Puerto Rican IDUs and found that 54% of Mexican Americans compared to 59% of Puerto Ricans injected heroin on a daily basis. Cocaine alone was injected daily by 16% of Mexican Americans and 44% of Puerto Ricans, while "speedball" (a combination of heroin and cocaine) was injected daily by 16% of Mexican Americans and 43% of Puerto Ricans. Clearly, in both Hispanic subgroups heroin is the preferred drug of injection. However, Puerto Ricans showed much higher injection rates for cocaine alone and "speedballs". Daily use of non-injected drugs was ascertained since this also has implications for sexual risk-taking behaviors. Mexican Americans reported higher rates of marijuana use than Puerto Ricans, but Puerto Ricans showed higher rates of using alcohol (32% vs. 28%, respectively) and much higher rates for daily use of crack (9% vs 2%, respectively), and cocaine powder (13% vs 5%, respectively).

C. Prevalence of HIV Risk Behaviors Among Hispanic IDUs

In assessing frequency of injection among Mexican Americans and Puerto Ricans, Estrada (1991) found that Puerto Ricans were almost twice as likely than Mexican Americans to inject drugs four or more times daily (41% vs 21%, respectively) which is not surprising given their higher rates of injecting cocaine and "speedballs". On the other hand, significantly more Mexican Americans shared needles with two or more people than Puerto Ricans (74% vs. 58%, respectively).

In terms of needle hygiene, Puerto Ricans were more likely than Mexican Americans to have used a clean needle (22% vs. 16%) and to have cleaned their needle (72% vs 62%). It is also important to know the method used to clean needles. The author found that Mexican Americans were much more likely than Puerto Ricans to clean needles with water only (42% vs. 28%). Alternatively, Puerto Ricans were more likely than Mexican Americans to clean needles with bleach and water (12% vs 9%) or alcohol and water (7% vs 5%).

Sharing of drug injection equipment was also examined. Results showed than Mexican Americans were much more likely than Puerto Ricans to rent or borrow used needles, and to share the cooker/cotton and rinse water.

According to NOVAs Needle Risk Index, wherein one can assess the overall level of risk associated with certain needle risk behaviors as described above, 6% of Mexican Americans and 11% of Puerto Ricans could be classified in the low risk category, 26% of Puerto Ricans and 17% of Mexican Americans could be classified at intermediate risk, while 75% of Mexican Americans and 63% of Puerto Ricans could be classified at high risk due to their needle risk behaviors.

With particular reference to sexual risk behaviors, specific areas examined included having a history of STD's, number of sex partners, and frequency of condom use. Puerto Ricans were much more likely to report a history of an STD than Mexican Americans (29% vs. 18%); no differences were found in the number of sexual partners between Mexican Americans and Puerto Ricans; but Mexican Americans were less likely than Puerto Ricans to report always using a condom (7% vs. 14%, respectively). Overall, using NOVA's Sexual Risk Index, 47% of Mexican Americans and 58% of Puerto Ricans could be classified as low risk; 24% of Mexican Americans and 18% of Puerto Ricans could be classified in the intermediate risk category; and 29% of Mexican Americans and 24% of Puerto Ricans could be classified as high risk in terms of their sexual behaviors.

Variations in risk behaviors by race, controlling for city-type was ascertained by Friedman et al., (1992). In this study, the investigators classified 19 NADR project cities into three groups: multi-cultural--Black/White/Puerto Rican, which included Hartford, Jersey City, Newark, New York, and Philadelphia; multicultural-Black/White/Mexican-origin, which included Houston, Long Beach, San Antonio, San Francisco, and Tucson; and Biracial--Black/White, which included Cincinnati, Cleveland, Columbus, Dallas, Dayton, New Haven, New Orleans, Portland (Oregon), and Washington. The investigators found drug injection frequency to be highest among Puerto Ricans in the multicultural--Black/White/Puerto Rican cities, and highest among Mexican-origin IDUs in multicultural--BIack/White/Mexican-origin cities. Additionally, Puerto Ricans were more likely to borrow used needles and shoot up in a "shooting gallery" than others in the multicultural-Black/White/Puerto Rican cities. Mexican-origin IDUs were also more likely to borrow used needles than others in multicultural--Black/White/Mexican-origin cities. Moreover, Mexican-origin IDUs were less likely than others to use bleach or alcohol to decontaminate needles.

Of course, one of the major criteria in propagating the HIV epidemic in any group is the seroprevalence of HIV in that community or region. Recent data obtained at intake from out-of-treatment IDUs participating in the NIDA National AIDS Demonstration Research Program (NADR) shows striking variations among Puerto Ricans and Mexican American IDUs who were recruited: The seroprevalence rate among the Puerto Rican cohort was 45.8 compared to a seroprevalence of 2.33 among the Mexican American cohort (NIDA, 1992). Other researchers have also noted higher seroprevalance rates among Puerto Rican IDUs in all regions of the U.S. (Northeast, Midwest, South, West) then in Mexican-origin IDUs (Selik, et al., 1989).

The above data clearly demonstrates variations in the types of drugs used, needle risk behaviors, sexual risk behaviors, and seroprevalence rates among Mexican American and Puerto Rican IDUs. The implications of the data presented on the variation of risk in specific subgroups of Hispanics calls for a serious consideration of prevention models needed to adequately target these groups within the context of an epidemiological paradigm which takes these variations into account and develops prevention programs congruent with their level of need and potential impact in preventing new cases, intervening in high risk areas, and caring for those who have become infected.

II. HIV Prevention Models and Their Theoretical Underpinnings

A. Theoretical Models

Currently, four major theories guide the development of HIV prevention models: the Health Belief Model (Rosenstock, 1974; Becker et al., 1977); Social Learning Theory (Bandura, 1977,1986); the Theory of Reasoned Action (Fishbein and Ajzen 1975; Ajzen and Fishbein, 1980); and Stages of Behavior Change (Prochaska and DiClemente, 1983; Proschaska and DiClemente, 1986). Each of these theories attempts to explain behavior as a process involving cognitive and sociocultural expectations, as well as proposing potential points of intervention to reduce risk behaviors or maintain safer behaviors.

The Health Belief Model (HBM) asserts that individuals will generally not engage in preventive health behaviors unless they possess minimal levels of relevant health motivation and knowledge; view themselves as potentially vulnerable to the disease and the disease as threatening; are convinced of the efficacy and feasibility of the advocated health behavior; and see few difficulties in undertaking the recommended health behavior. Within this context it is also assumed that demographic, personal, structural, and social variables are capable of influencing the adoption of protective health behaviors. However, these variables are hypothesized to work primarily through their effects on the individual's health motivations and subjective perceptions rather than functioning as direct causes of health action.

Social Learning Theory (SLT) is based on the social parameters which influence the adoption of behaviors. Relevant concepts include operant learning through the observation of behaviors that others in the social group engage in, value, and condone; normative behaviors; social networks that influence the behavior of individuals in that network; and self-efficacy or the belief that one can actively and positively affect behavior through skills or attitudes acquired. Within this

framework one could also include the concept of "normative conflict" which impacts one's perceptions about oneself in comparison to group or cultural norms.

The Theory of Reasoned Action (TRA) argues that individuals consider the implications of their actions before they decide to engage or not engage in a given behavior. As such, this theory views a person's intention to perform or not perform a behavior as the immediate determinant of action. Further, a person's intention is a function of two basic determinants: attitudes toward the behavior, and the person's perception of the social pressures put on him/her to perform or not perform the behavior, termed the subiective norm. Additionally, attitudes are a function of beliefs. The beliefs that underlie a person's attitude toward the behavior are termed behavioral beliefs, and the beliefs underlying a person's subjective norm are termed normative beliefs. Variables such as age, socioeconomic status and ethnicity are considered external variables which influence the beliefs a person holds or the relative importance one attaches to the attitudinal or normative considerations. The primary hypothesis of the TRA is that behavior change is achieved by producing changes in beliefs. By influencing beliefs about the consequences of engaging in a certain behavior, change is produced in the attitude toward the behavior; and by influencing beliefs about expectations of specific referents (others in the social network/group) one can affect the subjective norm. As can be seen, the TRA is similar in many ways to Bandura's SLT with its emphasis on normative behaviors in a social group, but differs in its focus on the importance of beliefs in structuring attitudes toward a given behavior. The TRA takes one step further, however, in positing the effectiveness of interventions. According to the Theory, the effectiveness of an intervention depends on the extent to which it produces the desired changes in primary beliefs and on the degree to which the assumptions linking these beliefs to attitudes, subjective norms, intentions, and behavior are met.

The Stages of Behavior Change hypothesizes that behavior change occurs in a series of steps. These steps or stages are labeled precontemplative (no intention to change one's behavior), contemplative (long-range intentions to change behavior), ready for action (short-range intentions to change behavior), action (attempts to change behavior), maintenance (long-term consistent behavior change), and relapse (change did not take hold and the person regresses to former risky behaviors). This theory has been used to guide maximum points of intervention for behavioral risk reduction.

B. Congruence of Theoretical Concepts to Mexican American IDUs

Given the theoretical underpinnings of the four major theories guiding HIV intervention/prevention efforts, researchers and interventionists must review the key concepts within each of the theoretical formulations and assess the congruence these concepts have with Mexican American culture, in general, and the "tecato" (Chicano or Mexican American heroin addicts) subculture in particular.

In reviewing the key theoretical concepts within the Health Belief Model one can readily ascertain the cultural and subcultural congruence. For example, the concept of "vulnerability to disease" is not necessarily perceived similarly by Mexican Americans. Commonly used sayings suggests that disease, including HIV infection and AIDS, is the result of fate or other factors beyond the immediate control of the individual. Thus, concepts like "locus of control" may have very little utility in explaining the adoption of relevant health behaviors among Mexican Americans. In particular reference to tecato subculture, Mata and Jorguez (1988) cite the need to assess whether users were "controlled" (i.e., more cautious in drug use habits) or "uncontrolled" (i.e., less cautious in drug use habits) in relation to their interpersonal interactions. They found that controlled users will readily grasp the idea of vulnerability, but uncontrolled users have not internalized this concept. Further, perceived seriousness of a disease, especially AIDS, was viewed differently by controlled and uncontrolled users, with uncontrolled users minimizing the threat of contracting AIDS. While these concepts may not be congruent with the "locus-of-control" idea, it is important to ascertain the meaning of control given the tecatos lifestyle.

Another important consideration is the perception of risk. Almost all Mexican American IDUs, and IDUs generally, take many risks in their daily lives (Connors, 1992). To the extent that HIV risk behaviors (sharing needles, poor needle hygiene, unprotected sex) are placed within the same context as other risks, perceptions of HIV risk will vary within the drug using subculture, generally being lower in the overall hierarchy of risk taking behaviors.

Bandura's Social Learning Theory requires that researchers and interventionists assess issues of self-efficacy. However, there is no direct Spanish translation in meaning for this concept, or for that matter with the concept of empowerment. There is no word in Spanish for "assert" as in assertiveness training. The closest equivalent terms are "to affirm", "to defend" and "to sustain", which do not convey the concept of being able to assert oneself in the face of disapproval or opposition. Further, "mastery" translates into Spanish as domination and superiority, but its intended meaning in terms of controlling one's interactions with others does not apply. Within the tecato subculture, self-efficacy may mean getting through the day, copping drugs relatively easily, or shooting-up drugs with little hassle, or what Mata and Jorquez refer to as "alivianarse" (to get well/straight).

Alternatively, Social Learning Theory and its focus on normative behaviors and the influence of social networks on individual behavior may have great applicability to the tecato subculture. Mata and Jorquez (1988) found that tecatos rely heavily on their social networks for learning how to shoot-up, how to obtain drugs, avoiding arrest, information and resource exchanges and coping with the tecato lifestyle. Estrada's (1991) research documents the extent of needle sharing and the sharing of other injection equipment among Mexican American IDUs social networks (i.e., sex partners, running partners, friends). Clearly, the implication of these observations and findings is that by targeting the individual's social network for

HIV risk reduction, interventionists may see positive changes in risk behaviors among Mexican American IDUs. Further, as Connors (1992) has pointed out, HIV risk occurs in the context of social interactions and social behaviors. Another aspect of SLT that has potential relevance is the use of a role model to effect behavior change. For tecatos, hearing a prevention message from a former "carnal" or especially a "veterano" who has escaped from "Ia vida loca" may be particularly influential as a credible source of information. Another implication from SLT is normative conflict. When reviewed in the context of "Ia vida loca" this may highlight stresses in the individual and social group vis-a-vis tecato lifestyle and Mexican culture, which may be an additional source of stress further propagating HIV risk behaviors and other drug using behaviors.

The Theory of Reasoned Action also has some significant applications to Mexican American IDUs. Since most tecatos share a common belief system in reference to "Ia vida loca", i.e., certain lifestyle characteristics and patterns of social interaction, interventionists may attempt to modify "behavioral beliefs" and "normative beliefs" by focusing attention on the drug using social network in addition to a focus on the individual.

Stages of Behavior Change can be used to help guide the phases of a particular intervention by assessing the stage of cognitive readiness of the individual to engage in appropriate risk reduction alternatives and overall risk management.

III. Designing Culturally Competent Interventions

The foregoing section dealt with the major theoretical formulations used to structure interventions for HIV risk reduction, and their potential applicability to Mexican American IDUs and the general tecato lifestyle. In this section, I will discuss components needed to design culturally competent interventions for this specific at-risk population.

A. Cultural Sensitivity and Cultural Competency

In the design of an intervention specifically targeting Mexican American IDUs, or any racial/ethnic group, one must decide whether the overall program will be culturally sensitive or culturally competent.

The term cultural sensitivity as used in intervention paradigms refers to either a) translation of materials into Spanish, b) hiring bilingual/bicultural staff, c) delivering the intervention within the targeted community, d) recognizing cultural differences, or, e) all of the above. On the other hand, cultural competence, while having the same components as cultural sensitivity, differs by building the overall intervention on cultural constructs, normative/cultural beliefs, or the essence of the culture/subculture itself. Designing a culturally competent intervention is much more difficult because it requires a careful consideration of major theoretical components of risk reduction interventions within the context of the culture/subculture targeted. A key factor, then, is conceptual development of concepts and constructs as they pertain to Mexican American IDUs, and the application/implementation of these concepts to the proposed intervention.

B. Unique Concepts Specific to Mexican American IDUs

Given some of the unique characteristics and history of the tecato lifestyle, especially in relation to age and intergenerational differences, it is clearly important to assess the degree of acculturation, the impact of acculturative stress on risk taking behaviors, the type and density of drug using social networks in addition to supportive personal networks (e.g., social support), and their influences on beliefs, attitudes and behaviors related to drug use, drug cessation, and HIV risk.

The concept of acculturation, or the degree to which Mexican Americans are more "traditional" in their orientation to Hispanic culture or Anglo culture, has profound implications to the etiology of risk behavior as well as guiding the development of a culturally competent intervention. Level of acculturation is associated with friendship networks and the extent of interaction among these networks. Moreover, acculturation shapes the beliefs and attitudes one holds, which in turn affects behaviors one engages in. Through an assessment of acculturation and connectedness to a more traditional tecato lifestyle, we begin to have a method that can be used to examine drug using social networks, personal networks and risk taking behaviors, and then tailor an intervention to focus on these differences.

Acculturative stress has been defined by Berry (1980: p.21) as "behaviors and experiences which are generated during acculturation and which are mildly pathological and disruptive to the individual and his group (e.g., deviant behavior, psychosomatic symptoms, and feelings of marginality". In relation to Mexican American IDUs, there are two contexts in which acculturative stress might occur: the context in which a Mexican American lDU may choose between staying more "traditional" with respect to Mexican culture or assimilating with the dominant or Anglo culture; and the context of "hanging-out" or "shooting-up" with same ethnic friends or mixing with other racial/ethnic groups, i.e., being less traditional in terms of the tecato lifestyle. Both of these contextual stresses may lead to "normative conflict" as implicated by SLT and maladaptive behavior.

Following from the implications acculturation and acculturative stress have on the character of social networks, their values, beliefs, traditions, and engagement in risk behaviors coupled with a strong sense of "familiaism" within Mexican American culture, it is very important to include the concept of social support. As Mata and Jorquez (1988: P.9) note, "among (Mexican American) drug users, there was a continuing reliance on personal social support networks for learning to use drugs, obtaining drugs, and avoiding arrest; for information and resource exchanges; and for coping with the exigencies of illicit drug use".

With respect to building a culturally competent intervention targeting Mexican American IDUs, several core Hispanic cultural values must be included. These values have been described by Marin (1989) and Marin (1990) as "familiaism or the significance of the family to the individual, "collectivism" or the importance of extended family members and friends, "simpatia" or the context in which social interactions are smooth and positive, "personalismo" or the preference for relationships with same-ethnic members in a social group, and "respeto" or the need to maintain one's personal integrity and that of others. Additional, concepts like "carnalismo" or brotherhood "compadrazo" or the special relationships between godparents and, "machismo" or the positive aspects of caring for one's family should also be included. These concepts can be used to help guide specific components of a risk reduction intervention as well as assessing the extent of culturally competent implementation.

C. Toward a Convergent. Culturally Competent Model of Prevention

The preceding sections have attempted to articulate not only key concepts inherent in major theories of behavior change and prevention, but also to show the congruence of these concepts with Mexican culture on the one hand and tecato subculture on the other. Key cultural elements were also reviewed in order to develop culturally competent components to guide and anchor the intervention developed to target Mexican American IDUs. In this section, I will attempt to derive a culturally competent AIDS prevention model that is consistent with major thrusts in the area of prevention and behavior change but placed in the context of Mexican and tecato cultural nuances.

To begin with, a culturally competent intervention to reduce the risk of HIV transmission among Mexican American IDUs must, at minimum, include a theoretical model which links major constructs in a temporal sequence to the desired outcome. Second, these theoretical constructs must have cultural equivalence or congruence with tecato subculture. The most salient elements relevant to cultural competency should anchor major theoretical components, thus providing a sociocultural context that guides all aspects of the intervention.

The intervention model consists of four general stages (Figure 1) which may or may not be linear: assessment; salience; factors affecting the decision to take action; and, the action taken.

It must be stressed that the foregoing model cannot be delivered in isolation, focusing only on the individual or his codependent; rather, the intervention must focus on the individuals social and personal networks. Further, the social interactions between the interventionist and the targeted group must be guided by the cultural dimensions of respeto, personalismo, and simpatia. Interventionists should have a fundamental understanding of "Ia vida loca", and ideally should be former tecatos that can create an atmosphere of "carnalismo" in order to create unity among participants and their drug using social networks.

D. Implementation of Culturally Competent Interventions

As Marin (1989: p. 413) correctly asserts "Those who seek to develop successful prevention interventions for Hispanics need to consider using culture-specific values and norms, disseminating information in terms that are appropriate for Hispanics, and using channels of information that not only are perceived to be credible by Hispanics, but which are accessed by them."

The implementation of a culturally competent HIV prevention model must validate the clients cultural and subcultural experiences on the one hand, and encourage behavioral and attitudinal change within a sociocultural context on the other.

The first step in the implementation process is the hiring of culturally competent outreach and research staff. Outreach workers should be former tecatos of varying ages since research has shown that risk behaviors, drug use, and related attitudes are age-dependent. Further, communication is better facilitated when same ethnic and age peers are employed to conduct outreach. Research staff also need to be culturally competent because the interpretation of results derived from quantitative analyses cannot be isolated from the cultural nuances which formulate the information provided by the respondent. Cultural competence is extremely important when the research utilizes a qualitative approach to examine the context and meaning of specific actions and attitudes in the drug using subculture.

The second step in the implementation of a culturally competent intervention is to stress the importance of how interaction is facilitated. In this respect, communication and interactions with clients must be guided by the cultural concepts of respeto, simpatia, compadrazo and carnalismo. Training staff in more culturally appropriate methods of communication and interaction, including culturally appropriate "body language", will assist outreach workers and interviewers in working effectively with the targeted clientele.

The next step is designing the actual intervention. Although the previous section discussed some necessary components which should be included in a culturally competent intervention, I will address the content of the intervention here.

First, the HIV prevention message must be couched in terms that are readily understandable and acceptable to Mexican-American IDUs. This message must at once validate their experiences and allow them to examine ways to change existing risk behaviors and attitudes without appearing to be judgmental or superimposing of "mainstream" values. The format of the intervention could be either group or individual focused. In either case, the importance of drug using social networks, personal supportive networks, and the family, including children, as inhibitors or facilitators of behavioral and attitudinal change must be preeminent. Cultural concepts like acculturative stress should be discussed, in addition to the characteristics of "la vida loca" that inhibit behavior change and other characteristics which can be modified to better facilitate change in the individual and his drug using social network.

As mentioned previously, the messenger--whether it is through video, radio or in-person, should be a former tecato who either changed his risk behaviors prior to becoming infected, and/or those tecatos who are now HIV+ who contracted the virus through needle or sexual risk behaviors.

E. Measurement and Evaluation Issues

Given the sociocultural context in which AIDS intervention/prevention paradigms must be placed, it is crucially important to consider measurement issues of key theoretical concepts and their indicators. In this regard, quantitative assessments must be derived from qualitative approaches to understanding how risk behaviors are manifested in the individual and his/her drug using social network.

Two qualitative methods which have been used extensively with Hispanic populations in general are focus groups and participant-observation. Both of these methods can elucidate the social and cultural contexts relating to the salience of theoretical concepts, their congruence with extant values and beliefs, and social network influences on beliefs, attitudes, behavioral intentions, and behaviors themselves.

Based on such qualitative information, quantitative measures can be derived. Acculturative stress, for example, can be measured based on areas of discussion gleaned from focus group formats. Already, there are numerous acculturation scales used, most of which are based on language use and/or characteristics of friendship networks. Cultural concepts like personalismo, familiaism, machismo, respeto, simpatia and carnalismo can be defined and elaborated in focus groups, with resulting indicators used for the measurement of these concepts. Further, subcultural orientations to concepts like risk, importance of health, perceived susceptibility and seriousness of the disease, and barriers and cues to taking action can be readily ascertained. Once major concepts and their indicators have been identified, they can be placed within a cultural competent intervention model.

Evaluation of the model delivered is the next major consideration. The key to the evaluation of a culturally competent model is process evaluation, i.e., assessing the degree of fit between the intervention delivered and the cultural aspects which guide and influence its delivery to the targeted subgroup. Cultural aspects must be clearly defined and unambiguously measured. Staff must be assessed for their cultural and linguistic abilities in addition to their knowledge about nuances of the tecato lifestyle and its sociocultural influences on risk behaviors. The temporal sequencing of specific aspects of the intervention must be examined in light of the cognitive stages of the participants by using the Stages of Behavior Change Theory coupled with an assessment of the individual's drug using social network and its readiness for change and potential diffusion in addition to assessing the individuals supportive networks.

The impact of the intervention on reducing HIV risk can only be assessed if the intervention was delivered/implemented as planned. The focus, then, should not be solely on outcomes, e.g., reduced needle and/or sexual risks, but also on the evaluation of the overall conceptual model to gain an understanding of the influences of concepts on one another and their individual or synergistic effects on

behavioral and attitudinal outcomes. For example, does level of acculturation have direct influences on risk behaviors, or does acculturation influence the characteristics of social networks that in turn influence risk behaviors? In other words, indirect and interaction effects should also be ascertained in the impact evaluation. Impact evaluation should also elicit changes in the tecatos drug using social network and personal networks.

IV. Future Directions

There are numerous implications of the present paper with regard to research, public health policy, and outreach involving Hispanic IDUs.

A. Research Implications

A thorough understanding of the sociocultural context in which Hispanic IDUs interact is integral to designing culturally competent interventions. This understanding can only be derived from qualitative research approaches (ethnographic field research, focus groups, participant observation). However, it cannot stop there. Qualitative information can readily lend itself to the development of quantitative instruments that can better assess key concepts relevant to risk behaviors and behavioral change.

Further research is needed in the area of conceptual congruence of major theoretical constructs and their application to Mexican American culture and drug using subculture to better inform applications and interventions for Hispanic IDUs. Such research could make a significant contribution to our understanding of equivalent concepts in English and Spanish and their subcultural applications.

Third, intervention research must move beyond "generic" models. While certain cultural concepts are similar across all the major Hispanic subgroups, it is the nuances of a particular subculture that may have profound effects on the risk behaviors one is trying to modify. We can no longer treat Hispanic IDUs as if they were similar. In fact, research results demonstrate marked variations among various Hispanic subgroups. These variations should be taken into consideration when designing culturally competent interventions.

Social network analysis, including measures of social support, should take precedence over individualized research thrusts, especially in relation to MexicanAmerican IDUs. Clearly, the inclusion of cultural concepts like "familiaism" and "carnalismo", which implicate the individual's social network, must be part of the investigation.

B. Public Health Policy Implications

In a time when funds are being cut-back for prevention programs, it becomes necessary to prioritize areas based on level of need. However, this should not be

done at the expense of other subgroups. The article by Selik, et al. (1989) is particularly disconcerting in this light. These authors proposed that more resources be allocated to Puerto Rican IDUs than to Mexican American IDUs solely on the basis of HIV seroprevalence; the prevalence of HIV risk behaviors was not addressed. The suggestion of these authors, who represent CDC, is counter to the basic tenets of primary, secondary, and tertiary prevention as used in Public Health. Given the high prevalence of risk behaviors among Mexican American IDUs, and the fact that many could be classified as high risk, it is clearly important to focus attention on primary and secondary prevention. Further, as Hispanics, we cannot afford to pit one subgroup against another: The HIV epidemic puts all Hispanics on a continuum of risk.

C. Outreach Implications

All too often outreach efforts are not coupled with or guided by a theoretical model of HIV prevention/intervention. The result is that outreach staff view research as an intrusion into "real life". The fact, however, is that outreach and research are two sides of the same coin. Both must be integrated in order to intervene effectively with Mexican American IDUs.

A second implication for outreach is the cultural competence of outreach staff as well as that of the interventionists. Both must know and apply cultural concepts, not only those that apply to Mexican culture, but additionally, those concepts that affect the lifestyles of tecatos. These individuals must embody the cultural nuances in a culturally competent intervention model since more often than not, outreach staff are the first contact most IDUs will have with the intervention delivered.

In conclusion, the derivation of culturally competent intervention models to reduce HIV risk among Mexican American IDUs must begin as soon as possible. Conceptual development must proceed from major theoretical formulations commonly used in HIV risk reduction programs, but must be placed within the context of specific Hispanic subcultural values, norms and patterns of interactions, as well as the drug using subculture.

References

Ajzen, I., and Fishbein, M.: Understanding Attitudes and Predicting Social Behavior. Prentice-Hall, Englewood Cliffs, New Jersey, (1986).

Bakeman, R., et al.: The incidence of AIDS among Blacks and Hispanics. J. Natl. Med. Assoc., 79: 921-928 (1987).

Bandura, A.: Social Learning Theory. Prentice-Hall, Englewood Cliffs, New Jersey, (1977).

Bandura, A.: Social Foundations of Thought and Action: A Social Cognitive Theory. Prentice-Hall, Englewood Cliffs, New Jersey, (1986).

Becker, M.H., et al.: Selected psychosocial Models and Correlates of Individual Health-related Behaviors. Med Care, 15 (suppl): 27-46 (1977).

Berry, J.W.: Acculturation as Varieties of Adaptation. In: Acculturation: Theory, Models and Some New Findings, A.M. Padilla (ed.) AAAS Selected Symposium 39, p. 9-25. Westview Press, Boulder, Co., (1980).

Centers for Disease Control: HIV/AIDS Surveillance Report, March 1992:1-18.

Connors, M.M.: "Risk Perception, Risk Taking and Risk Management Among Intravenous Drug Users: Implications for AIDS Prevention", Soc. Sci. Med., 34(6):591-601,(1992).

COSSMHO: HIV/AIDS-The Impact on Hispanics in Selected States. Washington, D.C., 1991.

Curran, J.W., et al.: Epidemiology of HIV infection and AIDS in the United States. Science 239: 610-616, (1988).

Estrada, A. L.: Behavioral Epidemiology of HIV Risks Among Injection Drug Users: A Comparative Assessment. Paper presented at the HIV-AIDS Health Services Research and Delivery Conference, AHCPR, December 4-6,1991.

Fishbein, M., and Ajzen, I.: Belief, Attitude, Intention and Behavior: An Introduction to Theory and Research. Addison-Wesley Publ. Co., Redding, Massachusetts, (1975).

Friedman, S.R., et al.: Racial Differences in Injection-related risk Behaviors Among Street-recruited Drug Injectors in 19 United States Cities. Submitted J Acquired Immune Deficiency Syndromes, (1992).

Marin, G.: AIDS Prevention Among Hispanics: Needs, Risk Behaviors, and Cultural Values. Pub Health Reports, 104:411-415 (1989).

Marin, B.V.: AIDS Prevention in non-Puerto Rican Hispanics. In: AIDS and Intravenous Drug Use: Future directions for Community-based Prevention Research, edited Z. Amsel, R. Battses, and C. Leukefeld. NIDA Monograph 93, 35-52, (1990).

Mata, A.G., and Jorquez, J.S.: Mexican American Intravenous Drug Users' Needle-sharing practices: Implications for AIDS Prevention. In: Battjes, R.J. and Pickens, R.W. (eds.). Needle Sharing Among Intravenous Drug Abuses: National and International Perspectives. NIDA Research Monograph 80. Washington, D.C.: U.S. Government Printing Office, 40-58, (1988).

Moore, J.W.: Homeboys-Gangs, Drugs, and Prison in the Barrios of Los Angeles. Temple University Press, Philadelphia, Pa., (1978).

National Institute on Drug Abuse. Data from the National AIDS Demonstrations Research Program, (1992).

Peterson, J., and Bakeman, R.: AIDS and IV Drug Use Among Ethnic Minorities. J Drug Issues, 19: 27-37 (1989).

Prochaska, J.O. and DiClemente, C.C.: Stages and Processes of Self-Change of Smoking: Toward an Integrative Model of Change. J Consult and Clinical Psych., 51:390-395, (1983).

Proschaska, J.O. and DiClemente, C.C: Toward a Comprehensive Model of Change. In Treating Addictive Behaviors, W. Miller and N. Heather (eds.), Plenum Press, New York, N.Y., (1986).

Rosenstock, l.M.: The Health Belief Model and Prevention Behavior. Health Educ Monographs, 2: 354-386, (1974).

Selik, R.M. et al.: Birthplace and the Risk of AIDS Among Hispanics in the United States. Am J Public Health, 79: 836-839, (1989).